BIRD-BANDING A JOURNAL OF ORNITHOLOGICAL INVESTIGATION

VoL. 48, No. 3 SUMMER 1977 PAG•,S 193-300

INTRASPECIFIC AND INTERSPECIFIC SONG MIMESIS IN CALIFORNIA SONG SPARROWS

BY CAROL EBERHARDT AND LUIS F. BAPTISTA

INTRODUCTION

Geographical variation in song and song dialects have been popular research topics in the last two decades (reviews in Thielcke, 1969; Lemon, 1975). Song dialects in White-crowned Sparrows (Zonotrichia leucophrys) are well known. It has been found, how- ever, that although dialects exist for the subspecies Z. 1. nuttalli, pugetensis, and oria,tha (Marler and Tamura, 1962; Baptista, 1974, 1975, 1977; Orejuela and Morton, 1974; Baker, 1975), Alaskan White-crowned Sparrows (ga•belii) show no dialect variation (DeWolfe et al., 1974). Kroodsma (1971) and Borror (1959, 1975) found similar results for the Rufous-sided Towhee (Pipilo erythroph- thalmus)--although dialects are absent in eastern subspecies, birds in Oregon share similar themes. More recently, Mundinger (1975) reported distinct dialects for eastern House Finches (Carpodacus •exicanus). However, Bitterbaum and Baptista (in prep.) found sharing of syllables and groups of syllables in California House Finches but no sharing of entire songs.

The Song Sparrow (Melospiza •elodia) is a highly variable singer with a large repertoire of songs and syllables (Mulligan, 1963, 1966). Wheeler and Nichols (1924) concluded that there was regional variation in pitch, pronunciation, or both that would enable one to separate birds of one locality from those of another. Borror (1961) and Harris and Lemon (1972) concur with Wheeler and Nichols, reporting dialects in the Song Sparrow. However, Mulligan (1966) studying California populations reported variability so great with- in a single individual, that no similarity could exist between indi- viduals.

During the 1973 breeding season, we noticed what was apparently a high incidence of theme sharing in a population of Song Sparrows in southern California. We also recorded what appeared to be Song Sparrow imitations of local Wrentit (Chamaea fasciata) syllables. These data, plus the conflicting reports on the presence or absence of song dialects in the Song Sparrow, and intraspecific geographical variation as to presence or absence of dialects in the species listed above, prompted us to conduct a detailed study of Song Sparrow songs in our region during the 1974 breeding season.

193

194] C. Eberhardt and L. F. Baptista Bird-Banding Summer 1977

_•mT•ODS AND MATERIALS

The Study Area The study was conducted at Scholl Canyon Park, Glendale,

California. It consists of seven man-made terraces against a hillside in a regional park covered with trees, bushes, and grass. Circum- annual watering keeps this area green, permitting the occupancy of Song Sparrows and making this a roesic ecological island surrounded by xeric chaparral. Eight Song Sparrows were recorded from this area. One additional individual, whose territory was separated from the canyon by a street and was approximately 0.25 miles from the other birds, was also recorded.

Identification of Individuals Birds were not banded but were recognized by individually

unique song patterns and singing locations as in Harris and Lemon (1972). Birds were recorded as living on one of the seven levels, either on the left side (• 1), the middle (•2), or the right side (•3). For example, the Song Sparrow on the right side of Level I was recorded as LI-3. The bird from the other canyon was labelled as O-1.

Analysis of So•gs Recordings of songs were made from March to June 1974, using

a Uher 4000 Report 1C Tape Recorder with an Electro-Voice 635A Dynamic Omnidirectional Microphone mounted on a 24-inch parabolic reflector. Tape speed was 3.75 in/see, and recordings were analyzed on a Kay Electric Sonograph, Model 7029A, using the wide band setting.

Three spectrograms of each song type per individual were made. Thereafter, only variants of those song types detected by audition were graphed. A catalogue of syllables and notes was then made as in Thompson (1970), permitting comparisons between indi- viduals. Terms used in the description of songs are those of Mulligan (1966).

SYLLABLE AND SONG SHARING BETWEEN BIRDS

The Individual Song Songs of Song Sparrows are highly variable, but may be character-

ized by short introductory phrases, a central trill, and a terminal flourish of notes (Borror, 1965; Mulligan, 1966). The male uses an extensive repertoire of songs that consist of: (1) distinct song types, each composed of unique sequences of syllables, notes, and note complexes (Fig. 1), and (2) variations of the former arising from the addition or omission of various notes in each theme.

Repertoire size for this population ranged from five to 12 themes per bird (excluding two birds for which less than 20 songs were recorded), with the mean being 7.33 (Table 1).

Syllable Sharing Comparison of all the syllables and notes found in the population

(Figs. 2 and 3) with each individual's repertoire revealed extensive

Vol. 4S, No. a Song Mimesis i• Song Sparrows [195

Bird-Banding 196] C. Eberhardt and L. F. Baptista Summer 1977

,ds d, am ,.

FIGUnE 1. Eleven of the 12 song patterns sung by Song Sparrow 0-1.

sharing between birds. No syllable was shared by every bird, and some syllables were unique to an individual's repertoire. Birds on the terraced slopes shared from 85.2% to 100% of the syllables in their individual repertoires. Bird 0-1, from the neighboring canyon, shared only 64.1% of his syllables (Table 1). All syllables used by birds L1-1, L2-1, L2-2, and L5-1 were shared with other individuals in the population, although not all syllables were shared with each other. However, only one to nine song types •vere recorded for these individuals. It is conceivable that other song types containing unshared syllables were not used and were thus missed by the investigators.

Degree of syllable sharing between pairs of Song Sparrows was determined using the method of Harris and Lemon (1972). For example, of the 27 notes and syllables used by L1-3 and the 24 used by L1-1, 14 were found in both repertoires, giving a value of (14 -t- 14)/(27 -t- 24) X 100 = 54.9%. All birds •vere compared in this manner, yielding an average of 42.19 _ 18.53% (Table 2). Exclusion of L2-1 and L2-2, for which less than five song types were recorded, gives a mean value of 56.2 _ 7.89%.

Theme Sharing Identical sequencing of like syllables was found in the repertoires

of four birds (Fig. 4). Additionally, five birds shared themes that were almost identical, each theme differing slightly in one or two syllable types in the introductory or terminal portions. When these data are pooled, five birds shared from one to three themes with

Vol. 45, No. 3 Song Mimesis in Song Sparrows [197

I 2 • 4 5 6 7

12 I• f• 15 16 17 18 19 20 21

22 2• 24 2__5 26 27 28 29 '50 •! •2

- •

k r" I 1

0 secs 0.4

FIGURr 2. Catalogue of shared syllables and notes from Song Sparrow songs.

from one to three other individuals (Table 3; Fig. 4). Song pairs A1-A2 and D1-D2 (Fig. 4) are identical in sequencing and number of syllables. Pairs B1-B2 and C1-C2 are not exactly identical, variations being in syllable morphology rather than sequencing. Songs A1, B1, and C1 are from bird L4-3; A2, B2, and C2 are from bird LI-1, illustrating two birds sharing three themes.

IN•ERSrECIFIC MIMICRY

Armstrong (1963) points out that some arian species, such as the Mockingbird (Mimus polyglottos) and Gray Catbird (Dumetella carolinensis), mimic regularly whereas others, such as Meadow- larks (Sturnella spp.), mimic only closely related species. Still

198] C. Eberhardt and L. F. Baptista Bird-Banding Summer 1977

TABLE 2

Degree of syllable sharing between pairs of Song Sparrows (%)

Bird 0-1 LI-1 L1-3 L2-1 L2-2 L2-3 L4-2 L4-3 L5-1

0-1 -- 47.6 45.4 17.4 18.6 51.5 50.8 47.1 49.3

L1-1 -- -- 54.9 32.3 14.3 54.9 69.2 67.9 60.5

L1-3 -- -- -- 41.2 25.8 63.0 65.5 50.0 52.5

L2-1 .... 18.2 23.5 28.6 33.3 15.4

L2-2 ..... 25.8 18.8 18.2 8.7

L2-3 ...... 61.8 50.0 56.5

L4-2 ....... 63.2 68.1

L4-3 ........ 50.5

other species, such as the White-crowned Sparrow, have demon- strated no interspecific mimicry (Marlcr and Tamura, 1964; MarleE, 1970).

Mulligan (1966) cross-fostered Song Sparrows under Canaries and found no mimicry of the foster-parents' vocalizations. However,

9 I0 I.I 12 13 14 15 16

18 •9 L:• 21 22 25 24 25

o

FIG•TI•E 3. Catalogue of unshared syllables and notes from Song Sparrow songs.

Vol. 4S, No. 3 So•q Mimesis in Song Sparrows [199

m

o

D}

seconds

F•auRE 4. Shared themes from the Scholl Canyon population of Song Sparrows. Songs A1, B1 and C1 are from bird L4-3. Songs A2, B2 and C2 are from bird L1-1. The other themes are from four other birds. Compare A1 with A2, B1 with B2, etc.

T•_B•.E 3

Number of themes shared between pairs of Song Sparrows

Bird 0-1 LI-1 L1-3 L2-1 L2-2 L2-3 IA-2 IA-3 L5-1

0-1

LI-1

L1-3

L2-1

I2-2

L2-3

L4-2

L4-3

L5-1

1

1 1 1

1 1 3

200] C. Eberhardt and L. F. Baptista Bird-Banding Summer 1977

Kroodsma (1976) repeated Mulligan's experiment and found that Song Sparrows may copy some song elements from Canaries. Except for our Song Sparrow L2-2, for which only one song type was recorded, all the other birds in our study used what were apparently Wrentit syllables in their repertoires. Seven of these eight birds used mimetic syllables in one theme and one bird used them in two themes. Three of the eight individuals used mimetic syllables in the introductory phrases, whereas five birds used them later in the song (Fig. 5).

Yeoman (1975) studied Wrentit songs in the San Francisco Bay region of California and described them as beginning with an intro- ductory portion consisting of a series of notes followed by a trill. Wrentit songs in our study area conform to this description (Fig. 5, A to D). Yeoman (1975) distinguished 10 types of territorial songs used by Wrentits; her classification was based on the number, kind, and arrangement of notes comprising each song. She recog- nized five basic note types, one of which (note type 4) was shaped like a cane or inverted "J." There are several variations of this note: one of these was rounded or dome-shaped (see introductory notes in song type 7 of Yeoman, 1975), a second type was pointed at the apex and shaped like a Gothic arch (as in trill of song type 5 of Yeoman, 1975), and a third type was shaped like an inverted "u" with two squared-off corners (song type 5 of Yeoman, 1975). All three of these notes occurred in our study area (songs A, B, and C

FIGu• 5. Songs A to D are Wrentit themes from Scholl Canyon. Songs E to J are Song Sparrow songs containing Wrentit syllables. Different types of arrows in Songs A to D indicate three different introductory syllable types used by Wrentits. The same arrows in Songs ]• to J illustrate the same Wren- tit syllables in Song Sparrow songs. Song G was recorded during our pilot study in 1973. The rest of the Song Sparrow songs were recorded in 1974.

Vol. as, No. 3 Sot•g Mimesis 6• So•tg Sparrows [201

respectively in Fig. 5), and all three of these notes appeared in the local Song Sparrow songs (compare songs A and I, B and G, C and H, Fig. 5).

Armstrong (1963:73) cautioned that calls of one species may fortuitously resemble those of other species. Could the similarities in the Wrentit and Song Sparrow syllables be a result of chance? This might indeed be the case if matching was found in only one syllable type. However, since resemblances were found in three syllable types, we would favor interspecific mimesis as an expla- nation for the similarities rather than convergence. Baptista spent many hours in the marshes of the San Francisco Bay of California (where Wrentits are absent) listening to Song Sparrow songs and did not hear these distinct introductory syllables.

DISCUSSION

Mulligan (1966) and Borror (1965) found a few cases of Song Sparrows sharing whole themes. However, Harris and Lemon (1972) found no instances of identical songs sung by two birds. Five Song Sparrows in our study shared themes with others in the population. Syllables and notes that comprise songs in our California popula- tion are also shared with other birds in the population. Syllable sharing is apparently more extensive for this community than in the one studied by Mulligan (1966) and somewhat more than in that studied by Harris and Lemon (1972). Song Sparrows from Quebec react more strongly to playback of songs from their home population than to songs from alien populations (Harris and Lemon, 1974). Harris and Lemon (1974) proposed that songs containing syllables frequently heard by the individuals elicited stronger responses than songs containing syllables with few or no familiar syllables.

Harris and Lemon (1972) have suggested that the apparent disparity between Mulligan's (1966) and Borror's (1965) and their own data might not be real because the authors emphasized dif- ferent parameters of song in their respective studies. Mulligan studied primarily patterns of timing, whereas Borror and Harris and Lemon treated variation in notes and combinations thereof. However, Mulligan (1966:15; and pers. comm.) actually looked for similar syllables in the repertoires of different birds and found very few. There is, therefore, geographical variation in the amount of intraspecific song or syllable mimesis in Song Sparrows as has been reported in White-crowned Sparrows, Rufous-sided Towbees, and House Finches (see Introduction).

There is, moreover, geographical variation in repertoire size. Borror's (1961) Maine population averaged 8.25 themes and birds from Quebec averaged nine themes (Harris and Lemon, 1972). In the San Francisco Bay area, Mulligan (1966:15) noted that no bird sang fewer than 10 themes. Thirteen birds that he studied in detail had repertoires ranging from 10-23 themes and averaged 15.8. Our southern California population averaged 7.33 themes (Table 1).

202] C. Eberhardt a•d L. F. Baptista Bird-Banding Summer 1977

The development of dialects requires isolation of the population, a means of transmitting the dialect between generations, and some source of variation that causes divergence between localities (Harris and Lemon, 1972; Nottcbohm, 1972; Baptista, 1975; Lemon, 1975). Additionally, in order to perpetuate dialects, males must continue to breed in the same area, and juveniles must not disperse over great distances. Marshall (1948) and Johnston (1961) found that California races of Song Sparrows are more sedentary and hold smaller territories than do races in other parts of the country. Moreover, Mulligan (1966) and Kroodsma (1976) have demonstrated song learning in the Song Sparrow. Why the higher incidences of theme sharing reduced repertoire size and the higher percentage of syllable sharing in the Southern California birds versus the northern Californian, Maine and Quebec birds studied by Mulligan, Borror, and Harris and Lemon?

Tompa (i• Mulligan, 1966:59) and Mulligan (in litt.) studied populations of Song Sparrows on Mandarte Island, British Colum- bia. They noted reduced repertoire size and high incidence of .sharing in this small and dense insular population. Song Sparrows m our study area live in a small ecological island (see Methods and Materials). We propose that perhaps the original colonizers brought with them a small repertoire of syllables and themes. Their des- cendants were, therefore, born into a relatively poor acoustical environment as compared with Song Sparrows raised in optimum Song Sparrow habitat supporting large numbers of males singing a large repertoire of syllables and themes. The increased amount of syllable and theme sharing in our birds is perhaps partially a reflection of a small pool of syllables and themes from which to mimic. Harris and Lemon (1972) also noted a high incidence of syllable sharing in birds holding adjacent territories versus birds that were more widely dispersed. Birds in our study area were packed into an isolated patch of habitat, probably increasing interactions between individuals during song development and thus increasing syllable and theme mimesis. Our data, may, there- fore, be an artifact of isolation in marginal habitat.

Nice's (1943) observations on song development in hand-reared captives lend some support to our hypothesis. Two birds were captured at about five or six days of age and hand-reared. They vicd with each other for dominance status and eventually sang six themes each, exact duplicates of each other, a phenomenon never otherwise observed.

Interspecific Mimicry in the Fringillidae

Imitation of conspecifics plays a large role in song learning in many species of birds (Marlcr and Mundinger, 1971; Nottebohm, 1972; Lemon, 1975). Interspecific mimicry has been less often documented. In laboratory studies, juveniles insufficiently exposed to songs of conspecifics may sometimes mimic interspecifically (Lanyon, 1960; Baptista, 1973). Species that naturally mimic in the wild arc apparently not as prevalent. Among fringillids, Euro- pean Greenfinches (Chloris chloris) have been found to freely

Vol. 4S, No. 3 So•g Mimesis in Song Sparrows [203

mimic other sympatric passtrines (Giittingcr, 1974). Some fringil- lids such as the House Finch (Baptista, 1972), Vesper Sparrow (Pooecetes grawineus) (Kroodsma, 1972), Oregon Junco (Junco oreganus) (Laidlaw Williams, in press), Chaffinch (Fri•gilla coelebs) (Thielcke, 1962; Knecht and Scheer, 1968), Chipping Sparrow (Spizella passerina), Bachman's Sparrows (Ai•ophila aestivalis) and Rufous-sided Towhee (Borror, 1968) are facultative mimics. Kroodsma (1973) and Baptista (1972) suggested that insufficient exposure to songs of adult conspecifics in nature might stimulate interspecific learning of songs.

Since mimicry in Song Sparrows is apparently rare, mimesis in this population of these emberizines merits an explanation. Perhaps because few conspecifics were present and few conspecific songs were heard, the founders and their progeny mimicked interspecifi- cally. Introductory phrases in Wrentit songs bear some resemblance to those in "pure" Song Sparrow songs in pitch, syllable duration and rhythm. These similarities may have prompted Song Sparrows to select Wrentit phrases as models for mimesis. Robinson (1975) recently studied Lyrebirds (Menura spp.) that regularly mimic a variety of species and found, similarly, that songs selected for mimesis were similar in frequency to conspecific songs.

SUMMARY

Variation in the songs of Song Sparrows in a California popula- tion wcrc studied. Most of the syllables and notes in a male's song repertoire wcrc shared with the other birds in the community. Some birds also shared from one to three whole themes with other individuals. Mimicked Wrentit syllables wcrc found in the intro- ductory or later phrases of the repertoires of eight of the nine birds. Song Sparrows arc geographically variable in their tendency to mimic intraspccifically, southern California populations sharing more syllables than do northern Californian, eastern and Canadian populations. It is proposed that increased syllable and theme sharing might bca reflection of a small population in marginal habitat exposed to a small pool of syllables from which to mimic.

ACKNOWLEDGMENTS

We thank Donald E. Kroodsma and James Mulligan for reading an earlier draft of this manuscript, for all their helpful comments, and for sharing with us their unpublished material on song de- velopment in Song Sparrows.

LITERATURE CITED

A•sF2o•. E. A. 1963. A Study of Bird Song. Oxford, Oxford University Press. B.•KE2, M. C. 1975. Song dialects and genetic differences in White-crowned

Sparrows ( Zo•otrichia leucophrys ). E•,ol•tio•, 39:266-270. B.•PTXSTA, L. F. 1972. Wild House Finch sings White-crowned Sparrow song

Z. Tierpsychol., 39:266-270. ß 1973. Song mimesis by a captive Gouldtan Finch. Auk, 90:891-894. ß 1974. The effects of songs of wintering White-crowned Sparrows on song

development in sedentary populations of the species. Z. Tierpsychol., 34: 147-171.

204] C. Eberhardt and L. F. Baptista Bird-Banding Summer 1977

ß 1975. Song dialects and demes in sedentary populations of the White- crowned Sparrow (Zonotrichia leucophrys nuttalli). Univ. Calif. Publ. Zool., 10S :1-52.

----. 1977. Geographical variation in song and dialects in the Puget Sound White-crowned Sparrowß Co•dor, in pressß

BORROR, D. J. 1959. Variation in the songs of the Rufous-sided Towhee. Wilson Bullß, 71:54-72.

1961. Intraspecific variation in passerine bird songs. Wilson Bull., 73: 57'78.

--. 1965. Song variation in Maine Song Sparrows. Wilson Bullß, 77:5-37. --. 1968. Unusual songs in passerine birds. Ohio J. Sci., 65:129-138. ----. 1975. Songs of the Rufous-sided Towhee. Co•dor, 77:183-195. D•:Wo•.FE, B. B., D. D. K.•s•.•, .•u L. J. PE¾•o•. 1974. Prominent. variations

in the songs of Gambel's White-crowned Sparrows. Bird-Banding, 45:224-252. GC•o•R, H. R. 1974. Gesang des Griinlings (Chloris chloris). Lokale Unter-

schiede und Entwicklung bei Schallisolation. J. Ornithol., ll$: 321-337. H•ms, M., .•u R. E. Lr•o•. 1972. Songs of Song Sparrows (Melospiza melodia):

individual variation and dialects. Ca•. J. Zool., $9:301-309. ß 1974. Songs of Song Sparrows: responses of males to songs of different

localities. Co•dor, 7•i:33-44. Jon•s•o•, R. F. 1961. Population movements of birds. Condor, •i3:386-389. K•:cn•r, S., .•u U. Scur:rm. 1968. Lautausserung und Verhalten des Azoren-

Buchfinken (Fringilla coelebs morelertl Pucheran). Z. Tierpsychol., •$ :115-169ß K•oous•5, D. E. 1971. Song variations and singing behavior in the Rufous-

sided Towhee, Pipilo erythrophthalmus oregangs. Condor, 73:303-308. •. 1972. Variation in songs of Vesper Sparrows in Oregon. Wilson Bullß,

$4:173-178.

ß 1973. Coexistence of Bewick's Wrens and House Wrens in Oregon. Auk, 90:341-352.

----. 1976. A reevaluation of song development in the Song Sparrow. Anim. Behar., in press.

L5•¾o•, W. E. 1960. The ontogeny of vocalizations in birds. In Animal Sounds and Communication, W. E. Lanyon and W. N. Tavolga, eds. Amer. Inst. Biol. Sci., 7:321-347.

Lr•o•, R. E. 1975. How birds develop song dialects. Co•dor, 77:385-406. NIsR•.•, P. 1970. A comparative approach to vocal learning: song development

in White-crowned Sparrowsß J. Comp. Physiol. Psychol., Monograph 71, 2. M.x•.•, P., 5•u P. C. Mv•u•o•. 1971. Vocal learning in birdsß In Ontogeny

of Vertebrate Behavior, H. Moltz, ed. New York, Academic Pressß Ms•r•R, P., 5•) M. T.•u•5. 1962. Song "dialects" in three populations of

White-crowned Sparrows. Condor, •i4:368-377. ß 1964. Culturally transmitted patterns of vocal behavior in sparrows.

Science, 146:1483-1486. M•su5•, J. 1948. Ecological races of Song Sparrows in the San Francisco Bay

region: Part I Habitat and Abundanceß Co•dor, $9:193-215. Mu•.ms•, J. A. 1963. A description of Song Sparrow song based on instrumental

analysis. Proc. XIII Intern. Ornithol. Congr., p. 272-284. ß 1966. Singing behavior and its development in the Song Sparrow, Melo-

spiza melodia. Univ. Calif. Publ. Zool., $1:1-76. Mu•)•o•, P. C. 1975. Song dialects and celonization in the House Finch,

Carpodacus mexicanus, on the east coast. Condor, 77:407-422. N•c•, M. M. 1943. Studies in the life history of the Song Sparrow II. The be-

havior of the Song Sparrow and other passerines. Trans. Linn. Soc., N.Y., 6:1-238.

No•noI•, F. 1972. The origins of vocal lem'ning. Amer. Nat., 106:116-140.

Vol. 48, No. 3 Song Mimesis in Son Sparrows [205

Om:JU•,A, J A, A•'•) M. L. MOSTON. 1975. Song dialects in several populations of Mountain White-crowned Sparrows (Zonotrichia leucophrys oriantha) in the Sierra Nevada. Condor, 77:145-153.

ROBINSON, F. N. 1975. Vocal mimicry and the evolution of bird song. Erau, 75: 23-27.

TU•E•,CI<E, G. 1962. Die geographische Variation eines erlernten Elementes im Gesang des Buchfinken (Fringilla coelebs) und des Waldbaumlaiifers (Certhia familiaris ). Vogelwarte, 21:199-202.

ß 1969. Geographic variation in bird vocalizations. In Bird Vocalizations, p. 311-339, R. A. Hinde, ed. Cambridge, Cambridge University Press.

Tuo•rsoN, W. L. 1970. Song variation in a population of Indigo Buntings. Auk, 87:58-71.

WuEF•,•, W. C., .•N•) J. T. N•CUOLS. 1924. The song of the Song Sparrow. Auk, 41:444-451.

YEOMAN, E. 1975. The selective effect of vegetation density on songs of three races of Wrentit (Chamaea fasciata) in central California. M. A. thesis, San Francisco State University.

Moore Laboratory of Zoology, Occidental College, Los Angeles, Cali- œornia 900•1. Received 9 October 1976, •cccptcd 4 April 1977.